Minimum and Array-like Actions

[beelerchris]

The current algorithm assumes env.action_space.low = -env.action_space.high and env.action_space.high[0] = env.action_space.high[i] for all i, which is not the case for all environments.

[beelerchris]

Please see the latest Pull request (#29) for changes I have made to correct this.

[sfujim]

Thank you for your effort! I'm mainly using this repo as a reference implementation for those who want to reproduce the paper on benchmark environments, which use symmetry in the action space. This is an additional level of detail for more specific use-cases, so I'd prefer to leave it as is.